Interfaculty Agreement With
ICDAS
Coding During Training Workshops Using Selected Patient Tooth Surfaces and Interproximal Radiographs
A. Parolia, C. G. Toh, L. L. Seow, S. Bhatia, F. C. Smales ABSTRACT
Introduction
This study aimed to evaluate the reproducibility and accuracy of the International Caries Detection and Assessment System (ICDAS) in both clinical and radiographic detection of caries as demonstrated by dental faculty during training workshops.
Methods
Data from two evaluation exercises were used in this study, each conducted after dental faculty had completed ICDAS online training and an intensive 3‐h workshop, including coding practice with images, extracted teeth and radiographs. For the clinical site session, 75 unrestored tooth surfaces from 28 subjects were coded by 21 dental faculty members, with magnified images aiding clinical surface location. For the radiographic session, bitewing radiographs of 64 proximal surfaces from 25 subjects were coded by 22 faculty members using ICDAS radiographic codes. After a 3‐h break, the coding procedures were repeated. Performance in coding reproducibility and accuracy (the latter against the gold standard) was assessed using quadratically weighted kappa statistics.
Results
Dental faculty coded clinical sites reproducibly, with 79.0% of kappa scores over 0.81 (excellent), and 21.0% between 0.61 and 0.80 (good), but equivalent percentages fell to 4.8% and 47.6% for coding accuracy (gold standard). The values for accuracy rose to 38.1% and 69.9% when ICDAS codes one and two were combined. With radiographs, 4.6% and 36.6% were found for reproducibility, and 4.5% and 45.5% for accuracy.
Conclusions
ICDAS demonstrated very good reproducibility and accuracy with clinical sites when codes 1 and 2 were combined. Faculty performance with radiographs indicated a need for improved methods of training in ICDAS radiographic coding.
Clinical Relevance
This study evaluated the reproducibility and diagnostic accuracy of ICDAS coding among dental faculty in clinical settings. These findings support visual ICDAS as a reliable clinical tool when one simple modification is made, producing excellent results with all types of specialist faculty. Integration of ICDAS into clinical teaching is supported, enhancing diagnostic precision in staging caries lesions and promoting uniformity in treatment planning. The study highlights the need to identify improved methods of training in ICDAS radiographic codes, which, if realized, will lead to further improvements in integrated oral healthcare.